Help me with hiss - Schalltechnik Growling Krizzly

[DIY Bass]

This is not a new build.  I have been using it for a while.  There is a bit of a story.

This is the schematic.  https://schalltechnik04.de/wp-content/uploads/2016/01/GK_04_Schaltplan.png

It was mounted on a pedal board, and had damage to the pre/post switch.  It sticks out the side of the box, and stuck a little too far over the side of the pedal board, and got broken in transit.  Oops.  Anyway, I don't often use the DI, and didn't have a replacement switch, so it stayed broken like that for a while until I could get a new switch.  I just kept on playing with.  Got the new switch eventually, and installed it.  Now I have quite a lot of hiss that was never there before. 

The obvious thing is remove the pre=post switch again, and I did that, but the hiss is still there.  It is the type of hiss you get with the treble all the way up.  Loud enough that I can't use the pedal. 

I have used an audio probe and traced the hiss back to the JFET (which is a J113).  There is no hiss on the gate, even with the audio probe quite loud.  The hiss is there on the drain.  I don't have any replacement J113s so I removed it, and built up the same part of the circuit (from R39 to C27) on a breadboard.  It worked fine - no hiss at all and just a bit of nice clean gain.  I figured the JFET was not likely to be the problem, so I installed it in the board again (I checked first and there was no hiss on any of the pads with the FET removed, but I didn't really expect there to be).  The hiss is still there on the drain, and through the rest of the circuit from then on. 

I have simulated the circuit on LTSpice (just the JFET stage, not the whole circuit) and the voltages I measure tally with the simulation (0v on gate, 1.6V on source and 6.5V on drain).  I am tempted to just replace all the rest of the components (resistors and capacitors) as I do have spares of all them, but before I do are there any other ideas or things I should check?  I am just worried I will replace them all and still not solve the actual problem.

[duck_arse]

opamps have way high cmrr, can reject a lot of noise on their supply. jfets have about not much, so if the supply has gorn bad, somehow, the oppies maybe don't care whereas the jfet barfs.

short answer - I'm no help.

[Clint Eastwood]

When you built it on the breadboard, did you use a different power supply?

[amz-fx]

Hum, Whine and Noise?

http://www.muzique.com/news/hum-whine-and-noise/

The Traco power converter can put out 100 mV p-p noise. Also, the datasheet says that the Traco should have no more than 47uF capacitance load, but the schematic shows 220u. Not sure how that will impact the circuit but it could contribute to the problem.

The series resistance of the 220u capacitors at the switching frequency of the Traco may need to be taken into consideration, and the original design might have selected a device that works well with the circuit. I don't see the switching frequency on the datasheet but I might have overlooked it as I was in a hurry.

Best regards, Jack

[DIY Bass]

When you built it on the breadboard, did you use a different power supply?

That is true.  I did use a bench power supply to power it.  There have been a few comments about potential power supply problems.  I will take a look at the supply rails on an oscilloscope and see what that looks like.

[DIY Bass]

Actually, I'll ask this one too.  If I was to use the bench supply to directly supply the power rails, with nothing plugged into the pedal power socket, would that risk breaking anything in the pedal power supply (I am thinking particularly about the Traco module).  If I can do that and essentially eliminate the pedal power supply then that would tell me a lot pretty quickly.

[amz-fx]

With a module like the Traco, you cannot get accurate readings on its performance unless it is driving a load.

Use a 1k resistor and a red LED from +15 to ground, and another resistor/LED from ground to -15v. This will load each rail with around 13ma, which will be enough.

regards, Jack

[DIY Bass]

No, I mean I want to power the rails with my external bench supply, with no power input into the Traco module.  That way if I still get noise problems I know the problem can't be with the Traco module.  I just want to be sure that the Traco module won't be vulnerable to voltage on the output pins with no voltage on it's input.  Otherwise I will have to desolder and remove it.

[Matthew Sanford]

I believe Jack meant to do that in addition to driving power from a different supply to remove issues caused by having no load for better readings to compare. I would assume that also means doing the same with the pedal power supply if initial readings were obtained without a load.

[Rob Strand]

No, I mean I want to power the rails with my external bench supply, with no power input into the Traco module.  That way if I still get noise problems I know the problem can't be with the Traco module.  I just want to be sure that the Traco module won't be vulnerable to voltage on the output pins with no voltage on it's input.  Otherwise I will have to desolder and remove it.

As a general rule, it's best not to drive power into the outputs.   There's always a risk of frying something and/or corrupting the results of the experiment.  I'd disconnect the input of the Traco's as well.  Noise can original from the switched input current of the converter.    If it's not connected it can't be damaged and can't cause any confounding issues. 

[amz-fx]

I agree that you should not feed voltage into the output of the Traco. It has a small transformer in it that probably cannot take much current and it would be easy to burn it out (and it's expensive).

regards, Jack

[mzy12]

I have this pedal in my back burner, so I don't know anything about potential hiss. One thing I'd like to see in pretty much any circuit that utilises a converter like the traco is something like a 10uH inductor on the input and both +v and -v output rails. I might add something like that on mine

[DIY Bass]

Thank you all for your help.  It's taken me a while due to lack of spare time, but things are looking up.  removed the Traco module and powered with +15/-15 from a bench supply.  It got worse.  Was getting a bad hum of the type that I associate with a broken joint or a cable plugged into the input jack with nothing connected.  That noise would vary a bit when I moved some of the off board wiring.  Now, when I made this pedal I was young and naive.  The kit came with a bunch of short pieces of hook up wire, which I had to join and heatshrink for longer runs.  It was some of these that were crackling and humming as they moved, although audio was still getting through and they measure very low ohms on conductivity.  Nevertheless I replaced all those joined wires and the hum and hiss have gone using the bench supply.  Next I just need to hook up the Traco module on a breadboard and power the unit from it.  If there's still no problem then I'll reinstall it.  If there is a problem then I'll replace it.  I think I will probably take Jack's advice and replace the filter caps with lower values as per the datasheet as well, just in case.

Thank you all for your advice and help on this one.

[Rob Strand]

removed the Traco module and powered with +15/-15 from a bench supply.  It got worse. 

Check continuity between the 0V line of the bench supply and grounds at various points in the circuit.

[DIY Bass]

removed the Traco module and powered with +15/-15 from a bench supply.  It got worse. 

Check continuity between the 0V line of the bench supply and grounds at various points in the circuit.

No, it was definitely the off board wiring.  I think it must have developed a small problem when I replaced the broken switch in the beginning - hence the hiss.  When I moved the board around to remove the Traco unit it obviously stress a wire even more to the point where it became very unhappy.  The wiring for the boost switch was what I noticed first.  Moving it around changed the volume of the noise a lot.  With all the wiring replaced it is now so quiet I thought at first I had broken it, but signal goes through on bypass and engaged, so it's not broken, just quiet :-)